Mining User Reviews for Key Design Features in Cognitive Behavioral Therapy-Based Mobile Mental Health Apps

Area of Science:

• Digital Health
• Human-Computer Interaction
• Psychology

Background:

• Cognitive behavioral therapy (CBT) mobile apps show promise in enhancing intervention effectiveness.
• Limited user-centered design guidelines exist for these rapidly proliferating mental health applications.
• This research addresses the need for evidence-based design principles for CBT apps.

Purpose of the Study:

• To identify critical design features of successful CBT-based mobile applications.
• To leverage user feedback from app reviews for design insights.
• To inform the development of more effective digital mental health tools.

Main Methods:

• Analysis of 24,902 four- and five-star app reviews.
• Preprocessing and bigram representation of review text.
• Latent Dirichlet Allocation (LDA) topic modeling and interactive visualization to identify design themes.

Main Results:

• Identified 86 distinct topics from app reviews, achieving high interrater reliability (Cohen's kappa = 0.86).
• Six primary design features emerged: mental health management/support, credibility, self-understanding insights, therapeutic tools, 'rescue sessions', and personal growth.
• These features collectively contribute to app effectiveness and user satisfaction.

Conclusions:

• The identified high-level design features provide a foundational framework for developing successful CBT mobile apps.
• These insights can guide developers in creating more user-centered and effective digital mental health solutions.
• This study offers actionable recommendations for optimizing CBT app design based on real user experiences.